| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.1892147587511826, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001892147587511826, | |
| "eval_loss": 2.5089292526245117, | |
| "eval_runtime": 19.6395, | |
| "eval_samples_per_second": 45.317, | |
| "eval_steps_per_second": 5.703, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005676442762535478, | |
| "grad_norm": 1.4552996158599854, | |
| "learning_rate": 3e-05, | |
| "loss": 2.7421, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.011352885525070956, | |
| "grad_norm": 1.1237053871154785, | |
| "learning_rate": 6e-05, | |
| "loss": 2.4469, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.017029328287606435, | |
| "grad_norm": 1.3720451593399048, | |
| "learning_rate": 9e-05, | |
| "loss": 2.3487, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.017029328287606435, | |
| "eval_loss": 2.2401163578033447, | |
| "eval_runtime": 19.4324, | |
| "eval_samples_per_second": 45.8, | |
| "eval_steps_per_second": 5.764, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02270577105014191, | |
| "grad_norm": 1.296636939048767, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.994, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.02838221381267739, | |
| "grad_norm": 1.1245951652526855, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.7114, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.03405865657521287, | |
| "grad_norm": 0.8451226353645325, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 1.5949, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03405865657521287, | |
| "eval_loss": 1.584686040878296, | |
| "eval_runtime": 19.5036, | |
| "eval_samples_per_second": 45.633, | |
| "eval_steps_per_second": 5.743, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.039735099337748346, | |
| "grad_norm": 1.0359187126159668, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 1.5339, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.04541154210028382, | |
| "grad_norm": 0.8172639608383179, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.5047, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0510879848628193, | |
| "grad_norm": 0.7217634320259094, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 1.4632, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0510879848628193, | |
| "eval_loss": 1.4855507612228394, | |
| "eval_runtime": 19.5232, | |
| "eval_samples_per_second": 45.587, | |
| "eval_steps_per_second": 5.737, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.05676442762535478, | |
| "grad_norm": 0.7281866669654846, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.403, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06244087038789026, | |
| "grad_norm": 0.7354586124420166, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 1.4402, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.06811731315042574, | |
| "grad_norm": 0.6528739333152771, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 1.4641, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.06811731315042574, | |
| "eval_loss": 1.4303480386734009, | |
| "eval_runtime": 19.4893, | |
| "eval_samples_per_second": 45.666, | |
| "eval_steps_per_second": 5.747, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.07379375591296121, | |
| "grad_norm": 0.6548705101013184, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 1.3606, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.07947019867549669, | |
| "grad_norm": 0.6964229941368103, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 1.4222, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.08514664143803216, | |
| "grad_norm": 0.7062962055206299, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 1.2746, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.08514664143803216, | |
| "eval_loss": 1.3992811441421509, | |
| "eval_runtime": 19.4911, | |
| "eval_samples_per_second": 45.662, | |
| "eval_steps_per_second": 5.746, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.09082308420056764, | |
| "grad_norm": 0.6801795363426208, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 1.4836, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.09649952696310313, | |
| "grad_norm": 0.7224637866020203, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 1.3836, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.1021759697256386, | |
| "grad_norm": 0.6693508625030518, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 1.3298, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.1021759697256386, | |
| "eval_loss": 1.3766961097717285, | |
| "eval_runtime": 19.5259, | |
| "eval_samples_per_second": 45.581, | |
| "eval_steps_per_second": 5.736, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.10785241248817408, | |
| "grad_norm": 0.7132231593132019, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 1.3428, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.11352885525070956, | |
| "grad_norm": 0.7401199340820312, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 1.4093, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.11920529801324503, | |
| "grad_norm": 0.7014519572257996, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 1.3732, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.11920529801324503, | |
| "eval_loss": 1.3639700412750244, | |
| "eval_runtime": 19.5187, | |
| "eval_samples_per_second": 45.597, | |
| "eval_steps_per_second": 5.738, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.12488174077578051, | |
| "grad_norm": 0.6970811486244202, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 1.3597, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.130558183538316, | |
| "grad_norm": 0.6701546311378479, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 1.2815, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.13623462630085148, | |
| "grad_norm": 0.714708149433136, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 1.3462, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.13623462630085148, | |
| "eval_loss": 1.3555618524551392, | |
| "eval_runtime": 19.5215, | |
| "eval_samples_per_second": 45.591, | |
| "eval_steps_per_second": 5.737, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.14191106906338694, | |
| "grad_norm": 0.6745234727859497, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 1.3576, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.14758751182592242, | |
| "grad_norm": 0.794704258441925, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 1.356, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.1532639545884579, | |
| "grad_norm": 0.7525967359542847, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 1.3508, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.1532639545884579, | |
| "eval_loss": 1.3505889177322388, | |
| "eval_runtime": 19.5169, | |
| "eval_samples_per_second": 45.602, | |
| "eval_steps_per_second": 5.739, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.15894039735099338, | |
| "grad_norm": 0.6943870782852173, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 1.2033, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.16461684011352887, | |
| "grad_norm": 0.6928514838218689, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 1.3266, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.17029328287606432, | |
| "grad_norm": 0.6602805256843567, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 1.2549, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.17029328287606432, | |
| "eval_loss": 1.3480561971664429, | |
| "eval_runtime": 19.5289, | |
| "eval_samples_per_second": 45.573, | |
| "eval_steps_per_second": 5.735, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1759697256385998, | |
| "grad_norm": 0.6837213039398193, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 1.3326, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.1816461684011353, | |
| "grad_norm": 0.7137513756752014, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 1.3411, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.18732261116367077, | |
| "grad_norm": 0.6947424411773682, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 1.357, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.18732261116367077, | |
| "eval_loss": 1.3478984832763672, | |
| "eval_runtime": 19.5217, | |
| "eval_samples_per_second": 45.59, | |
| "eval_steps_per_second": 5.737, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.5458351538176e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |